원문정보
초록
영어
Due to limited depth of field of machine vision cameras, multifocused image fusion is finding importance to produce a single image called fused image from various images of the same scene being imaged. To have focused images of all the objects in the scene, the fused image is formed by combining important features of various images. This in turn increases the importance of ability to assess the quality of the fused image more accurately. To be accurate, a typical image quality measure should be independent of image content, robust to noise, monotonic with respect to image blur and calculated with minimal computation complexity. In this paper, the performance of nine image quality measures were assessed through various experiments by applying image blur, adding image noise, changing image contrast and image saturation level. Experiments were also conducted on six sets of images to find the best image quality measure for multifocused image fusion.
목차
1. Introduction
2. Image Quality Measures
3. Evaluation of Image Quality Measures
3.1. Sensitivity to Image Blurs
3.2. Sensitivity to Image Noises
3.3 Sensitivity to Image Contrast
3.4 Sensitivity to Image Saturation
4. Multifocused Image Fusion
5. Conclusion
References